J Obes Metab Syndr 2020; 29(3): 237-239
Published online September 30, 2020 https://doi.org/10.7570/jomes20077
Copyright © Korean Society for the Study of Obesity.
GI Biopharma Inc., Westfield, NJ, USA
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
I read with great interest the recent paper by Walbolt and Koh1 regarding the effects of non-nutritive sweeteners (NNS) on obesity and type 2 diabetes. It reported that most prospective cohort studies found positive association between NNS and obesity or diabetes, while many of the randomized controlled studies showed neutral or even negative association. It provided some explanations such as the effect of the sweet taste of NNS on appetite and food intake or potential increased selection of NNS by people with obesity or diabetes in the prospective cohort that caused reverse causality. However, as discussed in the paper, increased weight gain with reduced food intake in livestock fed NNS suggested the existence of a true adverse effect of NNS on obesity and its closely-associated diabetes. Here, I would like to provide more explanations for discrepancies between the prospective cohort and randomized controlled studies using more detailed analyses.
Increased growth rate and weight gain in many livestock, including cows, pigs, sheep, and chickens, have been shown since the 1950s after treatment with a wide range of antibiotics at sub-therapeutic low dose but not at therapeutic high dose.2 For instance, 70 mg/animal/day may be used for growth promotion in cattle, while 6,600 mg/animal/day may be needed to treat an infection.2 This bell-shaped dose response was also seen in many other studies. Abou-Donia et al.3 treated groups of mice with 1.1–11 mg/kg/day sucralose for 12 weeks. The most pronounced significant increase in body weight was seen in the 1.1 mg/kg/day sucralose group, along with changes in gut bacteria. In addition, rats treated with 0.0005% (0.005 mg/mL) saccharin in drinking water for 6 weeks showed significant high blood glucose and lower food intake but higher weight gain;4 rats treated with 0.1 mg/mL (0.01%) saccharin for 7 weeks had increased blood glucose but no difference in body weight;5 while rats treated with 1.0%–7.5% saccharin showed a dose-dependent decrease in body weight.6
More importantly, this bell-shaped dose-response also was shown in randomized controlled clinical trials in humans. A study by Bueno-Hernández et al.7 showed that healthy volunteers receiving 48 mg sucralose daily for 10 weeks had significantly higher levels of blood insulin and glucose during an oral glucose tolerance test and reduced insulin sensitivity, while the 96 mg sucralose daily group failed to show such effects. The randomized controlled clinical trial by Romo-Romo et al.8 found that healthy adults ingesting 36 mg sucralose daily for 2 weeks had a significant decrease in insulin sensitivity. In contrast, a double-blind, placebo-controlled, randomized clinical trial showed no effect on insulin sensitivity or glucose tolerance in healthy subjects who consumed 1 g sucralose daily for 12 weeks.9 Together, these well designed clinical trials suggest the existence of a more potent adverse effect of low-dose sucralose than high dose.
How to explain the bell-shaped pattern of dose response? As discussed in the paper by Walbolt and Koh,1 the critical role of the gut microbiome on health has drawn increasing attention. Studies have well documented common changes in dysbiosis-associated health problems, such as decreases in abundance and diversity of gut bacteria, increased gut permeability, and systemic, low-grade inflammation and endotoxemia. Here, I propose that the bell-shaped response to NNS shares a similar mechanism to the low incidence of inflammatory bowel disease in both poor hygiene and germ-free conditions that may result from degradation of the protective mucus layer.10
The mucus layer is a crucial component of the gut barrier, and mucin, the key component of mucus, is a glycoprotein that contains a core peptide and many (up to 90% in weight) side carbohydrate chains. Thus, degradation of mucus requires both proteases, which mainly originate from the pancreas, to break down the peptide, and glycosidases, which come mainly from gut bacteria, to break down carbohydrates. Under conventional situations, digestive proteases are promptly inactivated by bacteria to prevent rapid degradation of protective mucus. A reduction in gut bacteria by NNS, such as saccharin or sucralose, may lead to impaired inactivation of digestive proteases that may work synergistically with bacterial glycosidases, resulting in accelerated degradation of the mucus layer. However, further reduction in gut bacteria may reach a point at which degradation will start to slow down due to great reduction in bacterial glycosidases. In addition, further reduction in gut bacteria would also reduce the amount and availability of bacterial lipopolysaccharides penetrating the body. These may result in decreases in endotoxemia and systemic inflammation and, thus, the potency of related adverse effects.
Randomized controlled studies showed potent adverse effects of sucralose on body weight and blood glucose at doses of 48 mg or 36 mg daily, an intake less than 1 mg/kg/day and a small fraction of the 5 mg/kg/day acceptable daily intake (ADI) set by the Food and Drug Administration in the United States and the 15 mg/kg/day ADI set by the Joint Food and Agriculture Organization and the World Health Organization Expert Committee in Food Additives. Thus, the peak of the most potent adverse effect may be much lower than the ADI, which may be closer to the dose consumed by the general population in the prospective cohorts. In contrast, the doses used in many randomized controlled trials may have far surpassed these most potent doses, which could contribute to discrepancies between prospective cohort and randomized control studies. NNS may have more potent adverse effects on health than are currently perceived. I suggest conducting more studies in this regard.
The author declares no conflict of interest.